Processing of high-dimensional and multi-scale data with support vector machines: Application to substorm forecasting.

Details Processing of high-dimensional and multi-scale data with support vector machines: Application to substorm forecasting. Processing of high-dimensional and multi-scale data with support vector machines: Application to substorm forecasting.

May 2001

adsabs.harvard.edu/cgi-bin/nph-data_query?bibcode=2001agusm..sm52a04g&link_type=abstract

American Geophysical Union, Spring Meeting 2001, abstract #SM52A-04

Physics

Geophysics

2447 Modeling And Forecasting, 3200 Mathematical Geophysics (New Field), 3210 Modeling

Scientific paper

Predictive capabilities of the data-driven geomagnetic substorm/storm models depend on the quality and amount of real-time data and on the algorithm used to extract generalized mappings. Availability of the high-resolution multi-scale data constantly increases. The best possible use of this observational information requires efficient processing and generalization of high-dimensional input data. The majority of advanced nonlinear algorithms can encounter a set of problems called "dimensionality curse". Neural networks (NN), one of the leading techniques for substorm/storm forecasting, are also sensitive to input dimensionality. A very promising algorithm that combines the power of the best nonlinear techniques and tolerance to very high-dimensional data is support vector machine (SVM). We have applied SVM to space science for the very first time to predict large-amplitude substorm events from solar wind data. We conclude that performance of SVM models can be comparable to or be superior to that of the NN-based models. The advantages of the SVM-based techniques are expected to be much more pronounced in future space-weather forecasting models, which will incorporate many types of high-dimensional, multi-scale input data once real-time availability of this information becomes technologically feasible.

Affiliated with

Also associated with

No associations

Rating

Processing of high-dimensional and multi-scale data with support vector machines: Application to substorm forecasting. does not yet have a rating. At this time, there are no reviews or comments for this scientific paper.

If you have personal experience with Processing of high-dimensional and multi-scale data with support vector machines: Application to substorm forecasting., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Processing of high-dimensional and multi-scale data with support vector machines: Application to substorm forecasting. will most certainly appreciate the feedback.

Rate now

Comments { 0 }

Profile ID: LFWR-SCP-O-1278323